Past studies of focal brain lesions, such as those arising from stroke or traumatic brain injury, focused on determining how specific behavioral deficits are related to the properties of the damaged tissue in a localization of function approach However, considerable evidence now suggests that focal lesions also affect the physiology of remote regions of the brain. Here, we investigate a novel view that the widespread consequences of a lesion may partly be predicted by large-scale network interactions of the damaged region. Specifically, some regions (connectors) have strong interactions across many different networks while other regions (system hubs) have strong interactions mainly within their own network. We hypothesize that these designations will predict the extent of impairments seen after brain lesions, providing important clinical and scientific insight into these system-levl effects. This hypothesis will be tested through two aims. Aim 1 examines the relationship between damage to connectors and hubs on behavioral performance across a number of domains in lesion patients. Aim 2 uses functional Magnetic Resonance Imaging to examine network interactions in these lesion patients' brains and healthy controls. This aim will examine interactions within pre-specified brain systems and whether system organization is itself altered by brain damage. Neural simulations will also be used to help interpret the findings. Preliminary evidence suggests that damage to connectors, but not to hubs, has pronounced impact on behavior across many different domains and network interactions in many different systems. We propose establishing this pattern in a large patient group with lesions to diverse locations to examine the generality of the preliminary findings and eliminate a number of potential alternative explanations for these effects. By examining disruptions after lesions, this research will shed light on the mechanisms by which networks are maintained in the healthy human brain and how regions with different network properties may contribute to behavior. Furthermore, these findings will provide important new insight into behavioral and brain deficits caused by lesions, improving the ability of clinicians to make prognoses for lesion patients and revealing new avenues to target for rehabilitation. The research aspects of this proposal will be complemented by a strong training regime for the applicant in (1) novel processing methods, (2) neuropsychological characterization of clinical populations, and (3) innovative network analysis techniques. These areas will allow the applicant to create a more sophisticated and multifaceted depiction of how brain networks contribute to function. The proposed research and training regime, in combination with the premier resources available in the Petersen lab at Washington University in St Louis, will position the applicant well for her future goal of becoming a independent investigator in the field, specializing in brain network physiology and it's connection to complex functions.